Connectivity Guaranteed Multi-robot Navigation via Deep Reinforcement Learning

Juntong Lin, Xuyun Yang, Peiwei Zheng, Hui Cheng
Proceedings of the Conference on Robot Learning, PMLR 100:661-670, 2020.

Abstract

This paper considers the multi-robot navigation problem where the geometric center of a multi-robot team aims to efficiently reach the waypoint without collisions in unknown complex environments while maintaining connectivity during the navigation. A novel Deep Reinforcement Learning (DRL)-based approach is proposed to derive end-to-end policies for the multi-robot navigation problem. In order to guarantee the connectivity during the navigation, a constraint satisfying parametric function (CSPF) is proposed to represent the navigation policy. Virtual policy extended environment (VP2E), an implementation framework of the CSPF is accompanied so as to make CSPF compatible with existing DRL techniques which rely on differentiable parametric functions. Both simulations and real-world experiments of a team of 3 holonomic robots are conducted to verify the effectiveness of the proposed DRL-based navigation method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v100-lin20a, title = {Connectivity Guaranteed Multi-robot Navigation via Deep Reinforcement Learning}, author = {Lin, Juntong and Yang, Xuyun and Zheng, Peiwei and Cheng, Hui}, booktitle = {Proceedings of the Conference on Robot Learning}, pages = {661--670}, year = {2020}, editor = {Kaelbling, Leslie Pack and Kragic, Danica and Sugiura, Komei}, volume = {100}, series = {Proceedings of Machine Learning Research}, month = {30 Oct--01 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v100/lin20a/lin20a.pdf}, url = {https://proceedings.mlr.press/v100/lin20a.html}, abstract = {This paper considers the multi-robot navigation problem where the geometric center of a multi-robot team aims to efficiently reach the waypoint without collisions in unknown complex environments while maintaining connectivity during the navigation. A novel Deep Reinforcement Learning (DRL)-based approach is proposed to derive end-to-end policies for the multi-robot navigation problem. In order to guarantee the connectivity during the navigation, a constraint satisfying parametric function (CSPF) is proposed to represent the navigation policy. Virtual policy extended environment (VP2E), an implementation framework of the CSPF is accompanied so as to make CSPF compatible with existing DRL techniques which rely on differentiable parametric functions. Both simulations and real-world experiments of a team of 3 holonomic robots are conducted to verify the effectiveness of the proposed DRL-based navigation method.} }
Endnote
%0 Conference Paper %T Connectivity Guaranteed Multi-robot Navigation via Deep Reinforcement Learning %A Juntong Lin %A Xuyun Yang %A Peiwei Zheng %A Hui Cheng %B Proceedings of the Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2020 %E Leslie Pack Kaelbling %E Danica Kragic %E Komei Sugiura %F pmlr-v100-lin20a %I PMLR %P 661--670 %U https://proceedings.mlr.press/v100/lin20a.html %V 100 %X This paper considers the multi-robot navigation problem where the geometric center of a multi-robot team aims to efficiently reach the waypoint without collisions in unknown complex environments while maintaining connectivity during the navigation. A novel Deep Reinforcement Learning (DRL)-based approach is proposed to derive end-to-end policies for the multi-robot navigation problem. In order to guarantee the connectivity during the navigation, a constraint satisfying parametric function (CSPF) is proposed to represent the navigation policy. Virtual policy extended environment (VP2E), an implementation framework of the CSPF is accompanied so as to make CSPF compatible with existing DRL techniques which rely on differentiable parametric functions. Both simulations and real-world experiments of a team of 3 holonomic robots are conducted to verify the effectiveness of the proposed DRL-based navigation method.
APA
Lin, J., Yang, X., Zheng, P. & Cheng, H.. (2020). Connectivity Guaranteed Multi-robot Navigation via Deep Reinforcement Learning. Proceedings of the Conference on Robot Learning, in Proceedings of Machine Learning Research 100:661-670 Available from https://proceedings.mlr.press/v100/lin20a.html.

Related Material